Re: RFC: next/finish/etc -vs- exceptions

[Pedro Alves <pedro at codesourcery dot com>]

[Reviewing bits and pieces instead of the whole patch at once]

On Friday 12 June 2009 21:44:30, Tom Tromey wrote:
> +/* A continuation callback for until_next_command. ?*/
> +
> +static void
> +until_next_continuation (void *arg)
> +{
> + ?struct thread_info *tp = arg;
> + ?delete_longjmp_breakpoint (tp->num);

This is broken, in that there's no guarantee that TP is still a
valid pointer here.

> +}
> +
> ?/* Proceed until we reach a different source line with pc greater than
> ? ? our current one or exit the function. ?We skip calls in both cases.
> ?
> @@ -1170,6 +1181,8 @@ until_next_command (int from_tty)
> ? ?struct symbol *func;
> ? ?struct symtab_and_line sal;
> ? ?struct thread_info *tp = inferior_thread ();
> + ?int thread = tp->num;
> + ?struct cleanup *old_chain;
> ?
> ? ?clear_proceed_status ();
> ?
> @@ -1205,7 +1218,18 @@ until_next_command (int from_tty)
> ?
> ? ?tp->step_multi = 0;??????????/* Only one call to proceed */
> ?
> + ?tp->initiating_frame = set_exception_breakpoint (frame);


> + ?old_chain = make_cleanup (delete_longjmp_breakpoint_cleanup, &thread);
> +
> ? ?proceed ((CORE_ADDR) -1, TARGET_SIGNAL_DEFAULT, 1);
> +
> + ?if (target_can_async_p () && is_running (inferior_ptid))
> + ? ?{
> + ? ? ?discard_cleanups (old_chain);
> + ? ? ?add_continuation (tp, until_next_continuation, tp, NULL);
> + ? ?}
> + ?else
> + ? ?do_cleanups (old_chain);

In sync execution mode, between that `proceed' and this do_cleanups, a
lot happens.  TP may exit, or the whole process for the matter.
So, after `proceed', any TP pointer is invalid.  Any other code
doing the same thing is equally broken.

-- 
Pedro Alves